Abstract: Microbial communities present new challenges for the development of statistics for biodiversity measurement. To understand the composition of microbial communities and the differences among communities across spatial locations, time points, or replicates, we present a new normalized measure for characterizing variability across multiple microbiome samples. The statistic relies on the diversity partitioning framework of the population-genetic statistic Fst, with samples playing the role of “populations” and taxa playing the role of “alleles” [1]. Its convenient mathematical properties--such as its commensurability across different numbers of taxonomic categories and different numbers of communities considered--allow users to compare its values between disparate data sets. Extensions incorporate phylogenetic similarity among taxa and spatial or temporal distances between communities. In a longitudinal analysis of gut microbiomes of healthy adults taking an antibiotic, we are able to both quantify the increase in temporal variability of microbiomes following the antibiotic course and to measure the duration of the antibiotic’s influence on microbial variability. The method is broadly applicable to many types of ecological data analysis.
[1] Morrison, M. L., Alcala, N., & Rosenberg, N. A. (2022). FSTruct: An FST-based tool for measuring ancestry variation in inference of population structure. Molecular Ecology Resources, doi.org/10.1111/1755-0998.13647.
